The influence of the geometry of the San Andreas fault system on earthquakes in California

1
The influence of the geometry of the San Andreas
fault system on earthquakes in California
Qingsong Li and Mian Liu Geological Sciences, 101
Geol. Bldg., University of Missouri, Columbia,
MO, 65211, USA (email qlpkd_at_mizzou.edu)
THE FINITE ELEMENT MODEL
MODEL RESULTS
INTRODUCTION
Predicted Energy Release and Seismicity
Predicted Velocity and GPS Velocity
At least three large (Mgt7.1) historical
earthquakes ruptured the San Andreas Fault (SAF)
since 1800, while more than a dozen Mgt7.0
earthquakes occurred outside the main trace of
the SAF. Most of the off-main-trace large
earthquakes were scattered in Southern
California, whereas in northern and central
California, earthquakes were clustered along the
main trace of the SAF. Here we use finite element
modeling to explore the stress pattern in
California. Our results indicate that the
distribution of earthquakes may be largely
related to the particular geometry of the SAF.
The strain energy is calculated by vertical
integration of energy release in plastic
deformation associated with failure of cells in
the upper crust, averaged over a time period of
8000 years.
The predicted surface velocities are comparable
to GPS velocities.
SEISMICITY IN CALIFORNIA
Predicted Energy Release and Active Faults
Predicted and Geological Slip Rates on SAF

• Model Parameters
• This is a dynamic finite element model the
  slip on the fault is calculated, not specified.
• Youngs Module is 8.75x1010 N/m2 and Poissons
  ratio is 0.25 for both crust and mantle.
• The model assumes plastic-viscoelastic rheology.
  The upper crust (blue color) is elastic
  viscosity for the lower crust and upper mantle
  (pink color) is 4.0x1019Pa s.
• The shear strength for the SAF fault elements is
  10 MPa during inter-seismic period and drops to 0
  MPa during co-seismic period.
• The Coulomb yield criterion is applied to the
  ambient upper crust. The Coulomb yield strength
  on optimal oriented plane is 45MPa. The effective
  internal frictional coefficient is 0.4 .

The predicted energy release pattern is
comparable with the diffuse off-main-trace
active faults in southern California. This
indicates that, similar to the earthquakes, some
of these active faults may have resulted from
stress evolution influenced by the particular
geometry of the SAF.
The predicted slip rates on the SAF in the model
are average values in a time period of 8000
years. The geological slip rates are from CGS
on-line database (http//www.consrv.ca.gov/CGS/rgh
m/psha/fault_parameters/htm/index.htm). We use
the sum of slip rates on several paralleling
branches of the SAF in northern California.
Maximum Shear Stress and Seismicity
CONCLUSIONS

• Modeling Procedure
• Stress loading by the relative plate motion and
  stress evolution during interseismic period is
  calculated at one-year time steps.
• 2. When stress reaches the strength at any cells,
  earthquake happens. The time step is reduced to
  1 second, and the cells are marked failed. In
  all the failed cells, plastic deformation is
  applied to keep the stress below the strength.
• 3. If failure of one cell causes stress in other
  cells to surpass their strengths, step 2 is
  repeated.
• 4. When no new cell fails, repeat step 1.

1. The geometry of the San Andreas fault plays a
   major role in the distribution of seismicity and
   active faulting in California. Along relative
   straight segments of the SAF, stress and energy
   release are concentrated near the main-traces of
   the SAF. Conversely, the bended SAF in southern
   California causes significant off-main-trace
   stress buildup. This helps to explain the diffuse
   seismicity including the large earthquakes
   outside the SAF in southern CA.
2. The dynamic model developed here involves much
   less assumptions than traditional kinematic
   models. It provides a more realistic simulation
   of earthquake cycles, stress evolution, and
   active crustal deformation over large regions.

The predicted maximum shear stress. The region
with high maximum shear stress (warm color) is
consistent with concentrations of off-main-trace
seismicity in southern California.
Most of the off-main-trace large earthquakes were
scattered in Southern California, whereas in
northern and central California, earthquakes were
clustered along the main trace of the San Andreas
Fault. The seismicity data are from NEIC catalog,
including Mgt6.0 earthquakes from 1800 to present.
Acknowledgements
We thank Eric Sandvol and Francisco Gomez for
helpful discussions. Huai Zhang helped
parallelizing the finite element codes for this
work.